Application of Godfrey Paul Armstrong et al, 280 F.2d 132 (C.C.P.A. 1960)Annotate this Case
As Modified July 20, 1960
Kenyon & Kenyon, New York City, Hercules Powder Co., Lawrence, Kan., Clinton F. Miller, Wilmington, Del., Solon B. Kemon, Kemon & Palmer, Washington, D. C. (Richard K. Parsell, New York City, of counsel), for appellants.
Clarence W. Moore, Washington, D. C. (Raymond E. Martin, Washington, D. C., of counsel) for Comr. of Patents.
Before WORLEY, Chief Judge, RICH, MARTIN and SMITH, Judges, and FRANCIS L. VAN DUSEN, Judge.1
This appeal is from a decision of the Board of Appeals dated October 28, 1957 and two decisions of that board dated September 30, 1958, affirming the final rejection of all of the claims remaining in application Serial No. 320,336, filed November 13, 1952, entitled "Manufacture of Peroxidic Compounds." That application is stated to be a division of application Serial No. 15,954, filed March 19, 1948, now U. S. Patent 2,632,772, issued March 24, 1953.
The specifications of those applications as filed for purposes of this appeal, are identical. The invention claimed in this case is directed to an anhydrous process of producing cumene hydroperoxide from cumene (isopropyl benzene). The anhydrous or homogeneous liquid phase process for producing cumene hydroperoxide is broadly distinguished from the aqueous or alkali aqueous phase oxidation by the absence of substantial quantities of water in admixture with the cumene. In both processes molecular oxygen as pure oxygen or air is maintained in contact with the liquid reaction mixture, and under proper operating conditions it reacts with the cumene to form the desired hydroperoxide. Present in the reaction mixture is an alkaline substance which functions to take up or react with undesirable carboxylic acid by-products. The specific alkaline substances claimed in this case are the alkali metal carbonates and more particularly sodium carbonate. The most advantageous temperature range is stated to be 70° C. to 150° C.
Of the claims on appeal, claim 3 is representative of claims 1 to 5, and claim 8 is representative of claims 7 and 8. They are:
"3. In a process for oxidizing cumene to cumene hydroperoxide in liquid phase with elemental oxygen as oxidizing agent, the improvement which comprises providing and maintaining solid sodium carbonate in contact with said cumene and employing temperatures in the range from above 90° C. to about 130° C.
"8. In a process for oxidizing cumene to cumene hydroperoxide in liquid phase with elemental oxygen as oxidizing agent, the improvement which comprises effecting said oxidation in homogenous [sic] liquid phase in the presence of alkali metal carbonate provided in the reaction mixture and employing temperatures in the range from about 70° C. to about 150° C."
Two rejections stand against claims 1 to 5 and a single rejection stands against claims 7 and 8. One rejection is based upon the alleged failure of the specification to support certain of the claimed limitations; the other is in the nature of a double patenting rejection. We shall consider them in the order in which we have named them.
Claim 3 had its genesis as claim 1 of U. S. Patent 2,681,9361 issued to one Joris. Claims 1, 2, 4 and 5 are patterned after that claim. Should claim 3 ultimately be found allowable to the present applicants, applicants' avowed intent is to provoke an interference with that patent.
The board affirmed the rejection of claims 1 to 5 as being unsupported by the specifications of both the parent2 and the instant applications as well as being fully met by Belgian patent No. 496,995.3 If those specifications do support the appealed claims, then the effective filing date of the present application (March 19, 1948) is clearly prior to the effective date of the Belgian patent and it is overcome as a reference against the claims on appeal.
The specific point of disagreement between the Patent Office and the appellants revolves around the phrases "providing and maintaining solid alkali metal carbonate in contact with said cumene" (claim 1) and "providing and maintaining solid sodium carbonate in contact with said cumene" (claims 2 to 5). It is the position of the Patent Office that neither the parent nor the instant specifications expressly or inherently supports the quoted limitations. With this conclusion appellants disagree. Sufficiency of support for the other limitations is not questioned by the Patent Office.
Since the specifications of the parent and the appealed applications (as filed) are identical for purposes of the issue under consideration, the discussion relating to the parent application will apply equally to the disclosure of the appealed application. If the parent application would have supported the claims here on appeal, so also then will the appealed application.
As to the homogeneous phase process for producing cumene hydroperoxide from cumene, the specification with respect to the alkaline substance which is to be present, states:
"When the process of the invention is carried out in the homogeneous liquid phase i. e. a liquid phase in which no immiscible diluent is present, the alkaline substances which are to neutralize the carboxylic acids present or formed during the reaction may be added in the form of a concentrated aqueous solution of water-soluble alkaline compounds such as alkali metal hydroxides.
"* * * The alkaline substances are preferably added in the form of concentrated solutions, for instance of alkali metal hydroxides in order to facilitate their introduction and their uniform distribution. The amount of water in the solution is advantageously limited so that at the reaction temperature it is evaporated with great rapidity. At any rate, it should be so limited that substantially no heterogeneous liquid phase is formed in the reaction mixture." [Emphasis ours.]
The disclosed alkaline substances which the application states may be used interchangeably in the oxidation process are alkali hydroxides and carbonates, including sodium carbonate, as well as alkaline earth oxides and hydroxides.
Example 4 of the parent specification is as follows:
"A reaction vessel equipped with a high speed agitator was charged with 150 cc. of purified isopropyl benzene to which 0.1 cc. of a 20% aqueous solution of sodium hydroxide had been added. The contents of the vessel were heated to 130° C. and oxygen of about 98% purity was introduced into the reactor. Absorption started within 15 minutes and the rate of absorption was 5.8 mol percent in the first hour of which 100% was converted into peroxide. In the second hour the oxygen absorption was 17.4 mol percent and the efficiency was 97%. The absorption of oxygen by the reaction mixture in the third hour was 20.6 mol percent, whilst the efficiency dropped to 93%. The oxygen introduction was then stopped before the absorption rate reached its previously ascertained maximum."
An affidavit by one Baird, submitted to show that solid alkali metal carbonate was present during the homogeneous phase oxidation described in the specification, was based upon the process of Example 4. Instead of using sodium hydroxide, sodium carbonate was substituted therefor. We agree with the board's understanding that if there is "appropriate language in the specification whereby one or more of the examples could be converted" to a carbonate process, it is unnecessary to have a specific example drawn to such a process to show the limitation in issue is inherently supported by the specification. In view of the equivalency which appellants teach for the various alkaline substances in their specification, and in view of the function which those substances play in the process, we feel that the conversion of Example 4 to a carbonate process from a hydroxide process is clearly within the scope of the parent disclosure.
There were several departures in the affidavit test from the steps of Example 4. Those were specifically pointed out by the Board of Appeals accompanied by the observation that there was no explanation for those changes in procedure. The test procedure included using quantities of cumene four times as great as that used in Example 4 as well as four times the equivalent amount of sodium carbonate instead of sodium hydroxide. No significance was attributed by the board to those changes.
The affidavit (in part) reads as follows:
"One liter indented Pyrex flask fitted with a high speed stirrer, reflux condenser, sparge tube, and thermometer. Heat was supplied with a Glas-Col mantle.
"Cumene (99.5%) 600 cc. "20% Aqueous Na2CO3 0.53 cc.
"The cumene was charged to the flask, heat and agitation applied, and 99.5% oxygen introduced at 300 cc./min. At 81° C. the 0.53 cc. of 20% NA2CO3 was added. After 10 minutes of heat-up the temperature reached 120° C. This point was taken as zero run time.
"Run Time % Cumene (hours) Temp. °C. O2 Input cc./min. Hydroperoxide 0 120 310 -- 0.25 129 350 0.20 1.0 131 500 5.22 2.0 132 510 25.6 3.0 129 510 44.2
"Samples were withdrawn at the end of each of the periods indicated above, analyzed for cumene hydroperoxide content by the iodine liberation method, and examined visually to determine whether there was a solid phase present. Each of the samples and the final product were found to contain a small amount of white solid crystalline material as a separate solid phase.
"The samples taken at the end of 0.25 and 2.0 hours and from the final product, identified as X-8098-47-1, X-8098-47-3, and X-8098-47, respectively, were given to William E. Fox of the Physics Group for examination of the solid phase thereof by x-ray diffraction. The solids were found by Mr. Fox to contain crystalline, anhydrous sodium carbonate in each case. No diffraction lines other than those for anhydrous, crystalline sodium carbonate were detected in the solids for many of the samples. Hence, the solids in each case were substantially entirely anhydrous, crystalline sodium carbonate."
The board's principal objection was to appellants' failure to follow the procedure of Example 4 exactly or to explain satisfactorily why they did not. As a general rule strict adherence to the steps described in the specification is essential if appellants wish to prove that those steps do in fact meet claim limitations. The reason for this is obvious, for after all it is the specification that is supposed to support claim limitations, not some other means contrived by affiants.
While the Board of Appeals is correct to the extent that departures from Example 4 do in fact exist, and as such the affidavits should receive the closest of scrutiny, we find that none of those departures affect the process in question, and therefore they are of no pertinence. In this case we are satisfied that Baird followed the steps of Example 4 sufficiently closely to render his affidavit of probative value with respect to the issue of the inherency of the limitation in issue. We find the deviations from that example to be inconsequential with respect to the question whether sodium carbonate is or is not present in its solid form and is in contact with cumene in that form during its oxidation to the hydroperoxide.
Contrary to what the board believed to be the case, the addition of aqueous sodium carbonate to the cumene after the temperature of the cumene has been raised to 81° C., instead of before heating of the cumene commences as required by Example 4, has no bearing upon the ultimate question here, as we see it. Neither does the addition of oxygen prior to reaching the reaction temperature.
Example 4 indicates that absorption of oxygen begins about 15 minutes after the reaction temperature is reached. The affidavit states that substantially no cumene hydroperoxide is present at the time the reaction temperature is reached. In both cases therefore, no or substantially no oxidation of cumene to its hydroperoxide occurs until the reaction temperature is reached. There is nothing of record to indicate that necessary intermediates form during the warm-up period, and, as a matter of fact, the first oxidation product of cumene appears to be its hydroperoxide.
Affiant states that within 25 minutes after the reaction temperature is reached, a "white solid crystalline material as a separate solid phase" is present in the reaction mixture. An affidavit of one Fox indicates that "the solids in each case were substantially entirely anhydrous, crystalline sodium carbonate," referring also to samples taken from the reaction mixture intermediate 25 minutes and three hours and ten minutes, the duration of the test period. We think that this showing is sufficient proof that appellants' specification teaches "a process for oxidizing cumene to cumene hydroperoxide in liquid phase with elemental oxygen as oxidizing agent, the improvement which comprises providing and maintaining solid sodium carbonate in contact with said cumene and employing a temperature of about 130° C." (Claim 5)
The Baird test was run at 130° C. The board said that
"* * * there is no adequate basis for Baird's conclusion that solid Na2Co3 would be present at temperatures of 70° C. to 150°. A single experiment at 130° C., well above the boiling point of water, would not establish similar results at 70° C., well below the boiling point of water."
That reason for the board's dissatisfaction with the affidavit obviously does not apply to claim 4 wherein the temperature range is 110° C. to 130° C. That leaves claims 1 and 2 (70° C. to 150° C.) and claim 3 (90° C. to 130° C.) for consideration in connection with that point.
It is interesting to note that the quantity of water present in Example 4 is at most 1 part of water to 1500 parts of cumene, very small indeed. The specification states as to all of the homogeneous phase reactions that "the amount of water in the solution is advantageously limited so that at the reaction temperature it is evaporated with great rapidity." [Emphasis added.] Since water evaporates at 70° C. as well as at 130° C., the quoted portion clearly suggests the use of less water to introduce the alkaline substance at lower temperatures or, alternatively, if the quantity used is the same as that used at higher temperatures, it should be so limited that it will evaporate speedily. For that reason we feel that the parent specification teaches how to vary the concentration of the carbonate solution of Example 4 to make it operative for temperatures as low as 70° C. while allowing the sodium carbonate to manifest itself in its solid form.
For the foregoing reasons we find that the board's conclusion that claims 1 to 5 are supported neither by the parent nor the instant specification is in error. Accordingly, the decision of the board with respect thereto is reversed. Furthermore, since the effective date of the appealed application antedates Belgian patent No. 496,995, the rejection based upon that patent is improper, and it too is reversed.
Since the point was raised, before we leave the question of inherency of support for the claims' limitations we might point out that we have found the limitation "providing and maintaining solid alkali metal carbonate in contact with said cumene" to require no specific mode of introducing the solid carbonate to the reaction mixture. No more is required than that the carbonate be present and that it remain there in contact with the cumene during the reaction. The affidavit shows that it is. Accordingly we need not comment upon any effect which the specification of the U. S. Joris patent,4 after which claims 1 to 5 are modeled, might have upon those claims.
Equally knotty is the rejection of all of the claims on claim 2 of the Conner et al. patent5 which is assigned to Hercules Powder Company, as is the appealed application, in view of any of Farkas et al.,6 Brewer7 or Lorand,8 the board finding the two former patents to be particularly pertinent.
Claim 2 of Conner et al. depends from claim 1. For purposes of simplified reading, the limitations of claim 2 have been inserted in claim 1 in brackets, and the portions for which the bracketed material is substituted has been deleted.
"Conner et al. Appellants' (Modified claim 1) Claim 8 "1. The process of oxidizing * * * In a process for oxidizing cumene to [cumene] to a high yield of * * * cumene hydroperoxide in liquid phase [ α, α -dimethylbenzy, hydroperoxide with oxygen as oxidizing agent, the (cumene hydroperoxide)] which comprises improvement which comprises affecting passing with intimate contact said oxidation in homogeneous liquid and under substantially anhydrous phase in the presence of alkali conditions an oxygen-containing gas metal carbonate provided in the reaction through a reaction mixture comprising mixture and employing temperatures essentially said * * * [cumene] in the range from about 70°C. to in liquid state at a rate of input such about 150°C. that at least the theoretical amount of oxygen is supplied at all times throughout the oxidation reaction, in the presence of a catalyst of the group consisting of alkali metal and alkaline earth metal oxides and hydroxides, and mixtures thereof, at a temperature between about 90° and about 145°C. for a time which is between 0.1 hour and a time which is no greater at any particular temperature than that defined by the equation loge A=11.672 — 0.0774t, where A=time in hours and t=temperature in said * * * [cumene] and said hydroperoxide having, respectively, the structural formulae * * *."
The board's rejection of the claims over Conner et al. claim 2, is in the nature of a double patenting rejection. It is, in effect, that Hercules Powder Company, the assignee of both the Conner et al. patent and the appealed application, has already patented in the Conner et al. patent that which the board alleges to be a single invention, thereby relinquishing any right it may have to that subject matter in another application, specifically the instant one.
Modified claim 1 of the Conner et al. patent adequately describes the claimed invention, and requires little amplification. The time-temperature relationship therein defined is specifically designed to give high yields of the desired hydroperoxides wherein at least 80% of the oxidized products are the hydroperoxides.
There are three limitations in the Conner et al. claim which find no counterpart in the appealed claims. They are:
1. The rate of input of the oxygen containing gas must be "such that at least the theoretical amount of oxygen is supplied at all times throughout the oxidation reaction."
2. There must be the specific time-temperature relationship.
3. Alkali metal and alkaline earth metal oxide and hydroxides are used as neutralization agents.
As to the first two limitations the board found in essence, that they were merely recitals of "suitable or optimum operating conditions and [that] such routine selections must necessarily be inferred in connection" with the appealed claims. This statement appears in the first board decision, wherein the rejection on Conner et al. was initially made. Under Rule 196(b), Patent Office Rules of Practice, 35 U.S.C.A.Appendix, appellants elected to renew prosecution before the Primary Examiner. Thereafter, that rejection was again considered by the board, this time in connection with affidavits submitted by Conner and the patent agent who had prosecuted the Conner et al. application. The substance of their decision on reconsideration of that rejection was the same as that of their initial rejection.
With respect to the third point, the equivalence of sodium carbonate and the claimed alkaline substances of Conner et al. the board relied on Farkas et al. and Brewer "which specifically teach the equivalence of alkali metal hydroxide and alkali metal carbonate in processes of this nature."
In arriving at our conclusion we have particularly concerned ourselves with the patented file of the Conner et al. patent and have carefully reviewed all of the prior art cited during its prosecution.
In the first Office Action in the Conner et al. prosecution, claim 1 was rejected as being too broad because the "rate of introduction of oxygen-containing gas is not specified and the proportion of oxygen in the gas is not specified."9 The examiner apparently was satisfied by its supplementation with a statement to the effect that the oxygen-containing gas is introduced at a rate of input "such that at least the theoretical amount of oxygen is supplied at all times throughout the oxidation reaction, * * *."
No corresponding rejection was made during the prosecution of the instant case which is some indication of the fact that the limitation is necessary to the Conner et al. claims and not necessary to those here on appeal. If it were merely the "expression of a suitable or optimum operating condition" [Emphasis ours] it would not have been necessary to the Conner et al. claims, as the examiner dealing with that case thought it was.
Several art rejections were made during the prosecution of the Conner et al. patent. Of the three Office Actions given, prior to the allowance of the claims therein, the first two carried rejections over a British Hall et al. patent,10 and over a patent to Brewer.11 The second action additionally rejected the claims over a U. S. Hall et al. patent,12 which patent was stated by the examiner to be "the U. S. duplicate" of the British Hall et al. patent, as well as on two additional patents.13 In the third action, dropped without comment were all of those rejections, and sole reliance was placed upon a U. S. Joris patent.14 One month and three days after applicants' response to that action a notice of allowance was mailed to them.
Conner and his co-applicants sought to avoid the British Hall et al. rejection, as well as the corresponding U. S. Hall et al. rejection, for the reasons that they dealt with the aqueous, rather than the claimed anhydrous phase oxidation of cumene, that it failed to show the claimed time-temperature relationship, the discovery of which, it was alleged, gave high hydroperoxide yields, and that it nowhere showed the calcium hydroxide catalyst of their application.
With respect to the Brewer patent it was principally argued that the hydrocarbons therein oxidized to hydroperoxides were so different from those claimed that the patent suggested nothing with respect to the oxidation of cumene. Additionally it was contended that the operating conditions of Brewer did not correspond to the claimed limitations.
As to the Lorand and Lorand et al. patents, applicants maintained that they were no more pertinent than the two Hall et al. patents. Both were argued to be directed to aqueous phase oxidation processes, neither, it was said, disclosing the operating conditions claimed whereby cumene hydroperoxide yields in excess of 80% of the oxidized products would be obtained.
As was stated earlier, all of those rejections were withdrawn without comment by the examiner as a result of his third Office Action. A rejection over the Joris patent,15 first made in the third Office Action, relied secondarily on Hall et al. to show the equivalence of the claimed and the Joris alkaline substances. The invention of the Joris patent is amply described in its claim 1 which reads:
"In a process for oxidizing cumene to cumene hydroperoxide in liquid phase with elemental oxygen as oxidizing agent, the improvement which comprises maintaining solid sodium bicarbonate in contact with said cumene, at reaction temperature in the range between about 60° C. and about 90° C."
Although they argued very briefly that the claims were distinguishable from the invention of the Joris patent, applicants swore back of that patent via a Rule 131 affidavit. Therefore, there was no final conclusion reached as to whether the Joris patent anticipated the Conner et al. claims. Notice of allowance of the claims followed that response by little more than a month.
As is readily observable from the synopses of applicants' arguments and the office actions in the Conner et al. file wrapper, applicants there with respect to all of the cited prior art, perhaps with the exception of the Joris patent, contended that the claims were patentable over that art because of basic differences in the nature of the processes, i. e., aqueous phase vs. homogeneous phase oxidation, other hydrocarbons vs. cumene, as well as because of the specific time-temperature relationship claimed. The claims were allowed over those arguments, the specific reasons for their allowance being unknown. Also considered important, by the examiner, during the Conner et al. prosecution was the rate of input of oxygen.
We feel that the file of the Conner et al. patent demonstrates that all of the limitations in its claims are important. The claims define unitary processes, all of the limitations discussed herein seemingly having been necessary to their allowance. Therefore, in the absence of some clear reason for holding that some of the features recited in those claims were not essential to their allowability, and we have found no convincing reasons in the record of this appeal, we find that they are not merely the "expression of suitable or optimum operating conditions." The invention claimed herein is broader and specifically different in some respects, and as such is a different invention. While both of them may be directed to the same end, that is not enough to make them the same invention. The rejection of claims 1 to 5, 7 and 8 as not being patentably distinct from claim 2 of the Conner et al. patent is reversed.
For the reasons stated the decision of the Board of Appeals is reversed.
United States District Judge for the Eastern District of Pennsylvania, designated to participate in place of Judge O'CONNELL, pursuant to provisions of Section 292(d), Title 28 U.S.C.
Issued June 22, 1954, filed August 3, 1950
Armstrong et al., U. S. Patent No. 2,632,772, March 24, 1953 (filed March 19, 1948)
Granted July 31, 1950, published November 3, 1950
Supra, footnote 1
U.S. Patent No. 2,632,774, March 24, 1953
U.S. Patent No. 2,430,864, November 18, 1947 (filed Jan. 30, 1945)
U.S. Patent No. 2,447,794, August 24, 1948 (filed January 30, 1945)
U.S. Patent No. 2,484,841, October 18, 1949 (filed May 14, 1947)
The limitation rejected as being too broad initially read:
* * * contacting said aromatic organic compound in liquid state with an oxygen-containing gas * * *.
British patent No. 610,293, October 13, 1948
Supra, footnote 7
Hall et al., 2,547,938, April 10, 1951
Lorand, 2,527,640, October 31, 1950; Lorand et al., 2,548,435, April 10, 1951
Joris, 2,577,768, December 11, 1951
Supra, footnote 14